Effectiveness of an algorithm in reducing the number of unnecessary ultrasound scans for deep vein thrombosis: an evaluation report

H’ng M, Loh SS, Earnest A, Wansaicheong G
Correspondence: Dr Martin H’ng Weng Chin, martin_hng@ttsh.com.sg

ABSTRACT
Introduction Patients with suspected deep vein thrombosis (DVT) pose a diagnostic dilemma to the Emergency Department (ED) clinician. This study aimed to implement a known algorithm incorporating the modified Wells criteria and D-dimer testing to guide the ED clinician, thus reducing unnecessary ultrasound scans (USS).
Methods Patients who presented to the ED between August 2008 and April 2009 with suspected DVT underwent Wells scoring. Those with scores < 2 were deemed unlikely to have DVT and underwent D-dimer testing first. Patients with scores ≥ 2 were regarded as likely to have DVT and underwent urgent USS. USS findings were tabulated as positive or negative/indeterminate for DVT. The latter group was followed up for one year to check whether DVT was missed during the initial USS.
Results 75 patients presented with suspected DVT and underwent USS. Of these, 14 results were positive and 61 were negative. 37 patients had Wells scores < 2, with three (8.1%) having DVT. Another 38 patients had Wells scores ≥ 2, with 11 (28.9%) having DVT. D-dimer testing was performed on 27 of the 75 patients. Those with DVT had higher average values compared to those without DVT (1.305 vs. 0.595 µg/ml). The majority of patients with raised D-dimer values had cellulitis, although three also had DVT (with values ≥ 0.99 µg/ml).
Conclusion We managed to reduce the number of unnecessary USS and increase the pick-up rate of DVT. A cut-off score ≥ 2 in our algorithm is suitable for use in the ED setting.

Keywords: D-dimer, deep vein thrombosis, modified Wells score, ultrasound scan, venous thromboembolism
Singapore Med J 2012; 53(9): 595–598